Journal of Experimental Botany, Vol. 60, No. 6, pp. 1663–1678, 2009doi:10.1093/jxb/erp034 Advance Access publication 13 March, 2009This paper is available online free of all access charges (see http://jxb.oxfordjournals.org/open_access.html for further details)

RESEARCH PAPER

Characterization of leaf apoplastic peroxidases andmetabolites in Vigna unguiculata in response to toxicmanganese supply and silicon

Hendrik Fuhrs1, Stefanie Gotze1, Andre Specht1, Alexander Erban2, Sebastien Gallien3, Dimitri Heintz4,

Alain Van Dorsselaer3, Joachim Kopka2, Hans-Peter Braun5 and Walter J. Horst1,*

1 Institute of Plant Nutrition, Faculty of Natural Sciences, Leibniz University Hannover, Herrenhauser Str. 2, D-30419 Hannover, Germany2 Max Planck Institute of Molecular Plant Physiology, Am Muhlenberg 1, D-14476 Potsdam-Golm, Germany3 Laboratoire de Spectrometrie de Masse Bio-Organique, IPHC-DSA, ULP, CNRS, UMR7178 ; 25 rue Becquerel, F-67087Strasbourg, France4 Institut de Biologie Moleculaire des Plantes (IBMP) CNRS-UPR2357,ULP, F-67083 Strasbourg, France5 Institute of Plant Genetics, Faculty of Natural Sciences, Leibniz University Hannover, Herrenhauser Str. 2, D-30419 Hannover,Germany

Received 7 November 2008; Accepted 26 January 2009

Abstract

Previous work suggested that the apoplastic phenol composition and its interaction with apoplastic class IIIperoxidases (PODs) are decisive in the development or avoidance of manganese (Mn) toxicity in cowpea (Vigna

unguiculata L.). This study characterizes apoplastic PODs with particular emphasis on the activities of specific

isoenzymes and their modulation by phenols in the Mn-sensitive cowpea cultivar TVu 91 as affected by Mn and

silicon (Si) supply. Si reduced Mn-induced toxicity symptoms without affecting the Mn uptake. Blue Native-PAGE

combined with Nano-LC-MS/MS allowed identification of a range of POD isoenzymes in the apoplastic washing fluid

(AWF). In Si-treated plants Mn-mediated induction of POD activity was delayed. Four POD isoenzymes eluted from

the BN gels catalysed both H2O2-consuming and H2O2-producing activity with pH optima at 6.5 and 5.5, respectively.

Four phenols enhanced NADH-peroxidase activity of these isoenzymes in the presence of Mn2+ (p-coumaric¼vanillic>>benzoic>ferulic acid). p-Coumaric acid-enhanced NADH-peroxidase activity was inhibited by ferulic acid

(50%) and five other phenols (50–90%). An independent component analysis (ICA) of the total and apoplastic GC-MS-

based metabolome profile showed that Mn, Si supply, and the AWF fraction (AWFH2O, AWFNaCl) significantly changed

the metabolite composition. Extracting non-polar metabolites from the AWF allowed the identification of phenols.

Predominantly NADH-peroxidase activity-inhibiting ferulic acid appeared to be down-regulated in Mn-sensitive

(+Mn, –Si) and up-regulated in Mn-tolerant (+Si) leaf tissue. The results presented here support the previously

hypothesized role of apoplastic NADH-peroxidase and its activity-modulating phenols in Mn toxicity and Si-

enhanced Mn tolerance.

Key words: BN-PAGE, cowpea, leaf apoplast, metabolome, manganese toxicity, phenolics, proteome.

Introduction

Manganese (Mn) in plants is an essential micronutrient

(Marschner, 1995). However, at supra-optimum supply Mn

readily becomes toxic to plants. Mn toxicity in crops is

a widely distributed plant disorder mainly on acidic and

insufficiently drained soils with low redox potentials thus

leading to high amounts of plant-available Mn (Horst, 1988).

In cowpea, Mn-resistant cultivars do not differ in Mn

accumulation from Mn-sensitive cultivars (Horst, 1980;

* To whom correspondence should be addressed: E-mail: horst@pflern.uni-hannover.deª 2009 The Author(s).

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/2.0/uk/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

by guest on September 29, 2015

http://jxb.oxfordjournals.org/D

ownloaded from

Fuhrs et al., 2008). Therefore, in this species Mn resistance

is regarded as Mn tolerance (Horst, 1983). Typical Mn

stress-induced toxicity symptoms in cowpea develop pri-

marily on older leafs as distinct brown spots located in the

leaf apoplast of the epidermis starting at the leaf base, then

spreading to the tip, followed by chlorosis, and, finally, leaf

shedding (Horst and Marschner, 1978b; Horst, 1982).

The brown spots consist of oxidized Mn and oxidizedphenolic compounds (Wissemeier and Horst, 1992). Hence,

the oxidaton of Mn2+ and phenols mediated by apoplastic

PODs was proposed to be a key reaction leading to Mn

toxicity (Fecht-Christoffers et al., 2006). Class III apoplastic

PODs (EC 1.11.17) belong to multigenic families (Passardi

et al., 2004) with various functions in plant growth (for

more information see Passardi et al., 2005). PODs are

polyfunctional enzymes that undergo two reaction cycles:the peroxidase–oxidase cycle (with NADH as substrate also

called NADH-peroxidases) resulting in H2O2 production

(Halliwell, 1978) and the peroxidase cycle (with guaiacol as

phenol substrate also called guaiacol-peroxidase) leading to

H2O2 consumption (Fecht-Christoffers et al., 2003a, b).

H2O2-producing POD activity was intensively studied with

respect to numerous exogenous factors like ambient pH

(Bolwell et al., 1995, 2001), phenol composition (Halliwell,1978; Fecht-Christoffers et al., 2006), and Mn2+ concentra-

tion in vivo (Yamazaki and Piette, 1963; Halliwell, 1978)

Fecht-Christoffers et al. (2006, 2007) investigated H2O2-

producing activity of apoplastic peroxidases of cowpea in

vitro and found that not only Mn2+ but also phenols are

required to induce NADH-peroxidase activity. Increasing

Mn concentrations in the leaf tissue and the AWF affected

the total apoplastic phenol concentration and composition.Crosswise combining of AWF metabolites with AWF

proteins from cultivars differing in Mn tolerance revealed

a significant effect on NADH-peroxidase activity. They

concluded that the apoplastic phenol composition and its

interaction with PODs are decisive in the development or

avoidance of Mn toxicity.

Silicon is a beneficial element for most plants (Epstein,

1999), and alleviates heavy metal toxicities, for example,aluminium and Mn toxicity. The alleviative effect of Si on

Mn toxicity was described for common bean and cowpea

(Horst and Marschner, 1978a; Iwasaki et al., 2002a, b),

cucumber (Rogalla and Romheld, 2002; Shi et al., 2005),

and pumpkin (Iwasaki and Matsumura, 1999). For cowpea,

Horst and Marschner (1978a) found that leaf Mn was more

evenly distributed in Si-treated cowpea plants. Horst et al.

(1999) demonstrated a reduction in apoplastic Mn concen-trations due to Si supply and concluded that Si changes

apoplastic Mn-binding properties, even though this could

only partly explain Si-mediated alleviation of Mn toxicity

(Iwasaki et al., 2002b). It was found that toxicity symptoms

and guaiacol-peroxidase activities were more closely related

to apoplastic Si concentrations than to apoplastic Mn

concentrations, indicating a more direct involvement of Si

nutrition in detoxification of apoplastic Mn.The work presented here specifically addressed the

hypothesis that the activities of specific apoplastic perox-

idases and their modulation by metabolites are decisive for

Mn toxicity and Si-induced enhanced Mn tolerance in the

Mn-sensitive cowpea cultivar TVu 91.

Materials and methods

Plant material

Cowpea [Vigna unguiculata (L.) Walp., cv. TVu 91] was

grown hydroponically in a growth chamber under con-

trolled environmental conditions at 30/27 �C day/night

temperatures, 7565% relative humidity, and a photon flux

density of 150 lmol m�1 s�1 photosynthetic active radiation

(PAR) at mid-plant height during a 16 h photoperiod. Aftergermination in 1 mM CaSO4 for 7 d, seedlings were

transferred to a constantly aerated nutrient solution with

four plants in one 5.0 l pot. The composition of the nutrient

solution was (lM): Ca(NO3)2 1000, KH2PO4 100, K2SO4

375, MgSO4 325, FeEDDHA 20, NaCl 10, H3BO3 8,

MnSO4 0.2, CuSO4 0.2, ZnSO4 0.2, Na2MoO4 0.05.

Silicon-treated plants (+Si) received Si in form of Aerosil

(Horst and Marschner, 1978a; chemically clean silicic acid,solubility in water: 0.6–0.75 mg l�1 or 20–26.5 lM). After

preculture for 14 d, the Mn concentration in the nutrient

solution was increased from 0.2 lM (–Mn) to 50 lM (+Mn)

for 4 d or 6 d. The nutrient solution was changed two to

three times per week to avoid nutrient deficiencies.

Extraction of water-soluble and ionically boundapoplastic proteins and metabolites

Apoplastic washing fluid (AWF) was extracted by a vacuum

infiltration/centrifugation technique according to Fecht-

Christoffers et al. (2003a, b). Leaves were infiltrated with

chilled dH2O by reducing the pressure to –35 hPa followed

by a slow relaxation. AWFH2O was recovered by centrifuga-

tion at 1324 g for 5 min at 4 �C. Afterwards, the same

leaves were infiltrated with chilled 0.5 M NaCl solutionand AWFNaCl was recovered as described above. Malate

dehydrogenase (MDH) activity in both AWF fractions

showed a cytoplasmic contamination of less than 1% (data

not shown). Until further analysis the AWF was stored

at –80 �C.

Quantification of toxicity symptoms

For the quantification of Mn toxicity symptoms, the density

of brown spots was counted on a 1.54 cm2 area at the base

and tip on the upper side of the second oldest middle

trifoliate leaf and calculated on 1 cm2 base.

Manganese analysis

Manganese in the bulk-leaf tissue was determined in thesecond oldest middle trifoliate leaf after dry ashing at 480 �Cfor 8 h, dissolving the ash in 6 M HCl with 1.5% (w/v)

hydroxylammonium chloride, and then diluting (1:10 v/v)

with double demineralized water. Apoplastic Mn concen-

trations were measured in 1:10 dilutions of the AWF. Both

1664 | Fuhrs et al. by guest on Septem

ber 29, 2015http://jxb.oxfordjournals.org/

Dow

nloaded from

measurements were carried out by optical inductively-

coupled plasma-emission spectroscopy (Spectro Analytical

Instruments GmbH, Kleve, Germany).

Silicon analysis

Monomeric Si concentration in the AWF was determinedaccording to Iwasaki et al. (2002a, b). AWF and a standard

solution (0–100 lg Si ml�1 AWF) were mixed with 250 ll ofstaining solution (1:1 mix of 0.08 M H2SO4 and 20 g l�1

(NH4)6Mo7O24.4H2O). After 30 min of incubation 250 ll offreshly prepared ascorbic acid (0.1 g 25 ml�1) and 250 lltartaric acid (0.85 g 25 ml�1) were added. Samples were

measured at k¼811 nm in a Microplate-Reader (lQuant,

BioTek Instruments, Germany).

Determination of the protein concentration in the AWFand AWF concentrates

The protein concentration in the AWF for the calculation

of specific enzyme activities was determined according to

Bradford (1976). The protein concentration of AWF

concentrates was measured for 1D BN-PAGE using the 2-DQuant Kitª (GE Healthcare, USA) according to the manu-

facturer’s instructions.

Determination of specific peroxidase activitiesin the AWF

For the measurement of H2O2-consuming guaiacol-peroxi-dase activities in the AWF, the oxidation of the substrate

guaiacol was determined spectrophotometrically at k¼470

nm (UVIKON 943, BioTek Instruments GmbH, Neufahrn,

Germany). Samples were mixed with guaiacol solution

(20 mM guaiacol in 10 mM Na2HPO4 buffer, pH 6) and

0.03% (v/v) H2O2. For calculation of enzyme activities the

molar extinction coefficient 26.6 l (mmol cm)�1 was used.

For the measurement of the H2O2-producing NADH-peroxidase activity in the AWF, samples were mixed with

MnCl2 (16 mM), p-coumaric acid (1.6 mM) and NADH

(0.22 mM). The NADH oxidation-dependent decline in

absorption at k¼340 nm was determined. For calculation of

enzyme activities the molar extinction coefficient 1.13 l

(mmol cm)�1 was used.

1D BN-PAGE of apoplastic proteins and PODactivity staining

For protein separation by electrophoresis under native

conditions, the proteins of the AWF were concentrated at

4 �C by using centrifugal concentrators with a molecular

mass cut-off at 5 kDa (Vivaspin 6, Vivascience, Hannover,

Germany). Running conditions were used according to the

manufacturer’s instructions.Proteins were separated via BN-PAGE according to

Jansch et al. (1996). Protein samples were combined with

Coomassie Blue solution [5% (w/v) Serve Blue G and 750

mM aminocaproic acid] and 10% (v/v) glycerol (100%).

Samples were loaded onto a native acrylamide gel with

a 4% (w/v) stacking gel and a 12% to 20% (w/v) gradient

separation gel. Electrophoresis was carried out at 100 V

and 6–8 mA for 45 min followed by 13 h at 15 mA (max.

500 V).

NADH-peroxidase activity in the gel was determined by

NBT staining to detect O��2 radicals or by DAB staining

(data not shown) to detect H2O2. The staining solution

finally consisted of 16 mM MnCl2, 1.6 mM p-coumaricacid, 0.22 mM NADH, and 2.5 mg ml�1 NBT in order to

detect O�2 radicals, that are proposed to be produced during

the NADH-peroxidase activity of PODs (Halliwell, 1978)

because a direct detection of H2O2 by DAB staining was

difficult due to the high gel background caused by Coo-

massie. Gels were stained for 30 min at room temperature.

The gels were afterwards soaked in 20 mM guaiacol (in 10

mM Na2HPO4) and 0.03% (v/v) H2O2 for 3 min to dectectguaiacol-peroxidase activity.

For preparative BN-PAGE guaiacol-peroxidase staining

was carried out only for a few seconds in order to reduce

enzyme damage by product–enzyme interaction.

Electroelution of specific POD isoenzymes for furtherphysiological characterization

Four POD isoenzymes (P1, P3, P5, and P6 in Fig. 3C) were

chosen for electroelution from BN gels that was carried outaccording to Wehrhahn and Braun (2002). POD isoenzymes

were cut from the gel and incubated for 30 min in cathodic

buffer [50 mM Tricine, 15 mM BIS-TRIS, 0.1 % (w/v)

Coomassie 250 G, pH 7 adjusted at 4 �C] and transferred

into the chambers of an electroeluter (CBS SCIENTIFIC,

Del Mar, USA). The gel pieces containing the POD

isoenzymes were filled into the electroeluter containing

elution buffer (25 mM Tricine, 7.5 mM BIS-TRIS, pH 7.0adjusted at 4 �C). Electroelution was carried out for 5 h and

4 �C at 350 V and 6–10 mA, using dialysis membranes

(Medicell, Kleinfeld) with a MWCO of 12–14 kDa under

constant buffer circulation (Econopump, Bio-Rad Labora-

tories, CA, USA). Until further characterization, eluates

were stored at –80 �C.

Determination of the pH optimum of the guaiacol-peroxidase and NADH-peroxidase activity of PODisoenzymes

For guaiacol-peroxidase measurements, 6 ll eluate was

mixed with guaiacol (20 mM) in 0.1 M succinate buffer

with the pH values 5, 5.5, 6, 6.5, and 7. The reaction was

started by adding 0.3% (v/v) H2O2. The increase in

absorption was measured at k¼470 nm using a Microplate

Reader. For calculation of enzyme activities the molar

extinction coefficient 26.6 l (mmol cm)�1 was used.

NADH-peroxidase activity measurements were made bycombining MnCl2, p-coumaric acid, and NADH in final

concentrations of 16 mM, 1.6 mM, and 0.66 mM, re-

spectively, with 7.5 ll protein eluate in 0.1 M succinate

buffer (as described above). The decline in absorption was

determined using a Microplate Reader at k¼340 nm. For

Apoplastic peroxidases and metabolites in manganese toxicity | 1665 by guest on Septem

ber 29, 2015http://jxb.oxfordjournals.org/

Dow

nloaded from

calculation of enzyme activities the molar extinction co-

efficient 1.13 l (mmol cm)�1 was used.

Determination of cofactor specificity for NADH-peroxidase activity of POD isoenzymes

The same experimental set-up as for the determination of

the pH optimum was followed using succinate buffer (pH

5.5). p-Coumaric acid was substituted by benzoic acid,

caffeic acid, chlorogenic acid, ferulic acid, gallic acid,

protocatechuic acid, syringic acid, vanillic acid, and p-

hydroxybenzoic acid in four different concentrations (1.66mM, 0.166 mM, 0.0166 mM, and 0.00166 mM) in the

measuring solution. In order to simplify this report, benzoic

acid as an aromatic carboxylic acid is termed as a phenolic

acid, too. For each phenol concentration specific extinction

coefficients were determined and used for enzyme activity

calculation (see Supplementary Table S1 at JXB online).

Determination of changes in NADH-peroxidase activityof POD isoenzymes as affected by combining differentphenols with p-coumaric acid

To detect the effects of different phenols on p-coumaric

acid-stimulated NADH-peroxidase activity of different iso-enzymes separated by BN-PAGE 0.166 mM p-coumaric

acid was combined with benzoic acid, caffeic acid, chloro-

genic acid, ferulic acid, gallic acid, protocatechuic acid,

syringic acid, vanillic acid, and p-hydroxybenzoic acid each

at a concentration of 0.0166 mM. All other factors were

kept as described for the measurement of cofactor specificity.

Activity was expressed as a percentage of p-coumaric acid

induced NADH-peroxidase activitiy. For each phenolconcentration, specific extinction coefficients were deter-

mined and used for enzyme activity calculation (see

Supplementary Table S1 at JXB online).

Mass spectrometric protein analysis anddata interpretation

Marked BN-PAGE bands stained for guaiacol-peroxidase

activity were cut and dried under vacuum. In-gel digestion

was performed with an automated protein digestion system,

MassPREP Station (Micromass, Manchester, UK). The gel

slices were washed three times in a mixture containing 25

mM NH4HCO3:acetonitrile (1:1, v/v). The cysteine residueswere reduced by 50 ll of 10 mM dithiothreitol at 57 �C and

alkylated by 50 ll of 55 mM iodacetamide. After de-

hydration with acetonitrile, the proteins were cleaved in the

gel with 40 ll of 12.5 ng ll�1 of modified porcine trypsin

(Promega, Madison, WI, USA) in 25 mM NH4HCO3 at

room temperature for 14 h. The resulting tryptic peptides

were extracted with 60% acetonitrile in 0.5% formic acid,

followed by a second extraction with 100% (v/v) acetoni-trile.

Nano-LC-MS/MS analysis of the resulting tryptic pep-

tides was performed using using an Agilent 1100 series

HPLC-Chip/MS system (Agilent Technologies, Palo Alto,

USA) coupled to an HCT Ultra ion trap (Bruker Daltonics,

Bremen, Germany). Chromatographic separations were

conducted on a chip containing a Zorbax 300SB-C18 (75

lm inner diameter3150 mm) column and a Zorbax 300SB-

C18 (40 nl) enrichment column (Agilent Technologies).

HCT Ultra ion trap was externally calibrated with

standard compounds. The general mass spectrometric

parameters were as follows: capillary voltage, –1750 V; dry

gas, 3.0 l min�1; dry temperature, 300 �C. The system wasoperated with automatic switching between MS and MS/

MS modes. The MS scanning was performed in the

standard-enhanced resolution mode at a scan rate of 8100

m/z s�1 with an aimed ion charge control of 100 000 in

a maximal fill time of 200 ms and a total of four scans were

averaged to obtain a MS spectrum. The three most

abundant peptides and preferentially doubly charged ions

were selected on each MS spectrum for further isolation andfragmentation. The MS/MS scanning was performed in the

ultrascan resolution mode at a scan rate of 26 000 m/z s�1

with an aimed ion charge control of 300 000 and a total of

six scans were averaged to obtain an MS/MS spectrum. The

complete system was fully controlled by ChemStation Rev.

B.01.03 (Agilent Technologies) and EsquireControl 6.1

Build 78 (Bruker Daltonics) softwares. Mass data collected

during LC-MS/MS analyses were processed using thesoftware tool DataAnalysis 3.4 Build 169 and converted

into .mgf files. The MS/MS data were analysed using the

MASCOT 2.2.0. algorithm (Matrix Science, London, UK)

to search against an in-house generated protein database

composed of protein sequences of Viridiplantae down-

loaded from http://www.ncbi.nlm.nih.gov/sites/entrez (on 6

March 2008) concatenated with reversed copies of all

sequences (23 478 588 entries). Spectra were searched witha mass tolerance of 0.5 Da for MS and MS/MS data,

allowing a maximum of 1 missed cleavage by trypsin and

with carbamidomethylation of cysteines, oxidation of

methionines, and N-terminal acetylation of proteins speci-

fied as variable modifications. Protein identifications were

validated when at least two peptides with high quality MS/

MS spectra (Mascot ion score greater than 31) were

detected. In the case of one-peptide hits, the score of theunique peptide must be greater (minimal ‘difference score’

of 6) than the 95% significance Mascot threshold (Mascot

ion score >51). For the estimation of the false positive rate

in protein identification, a target-decoy database search was

performed (Elias and Gygi, 2007).

GC-MS-based metabolite profiling

For GC-MS analysis, polar metabolite fractions were

extracted from 60 mg 610 % (FW) frozen plant material,

ground to a fine powder, with methanol/chloroform. The

fraction of polar metabolites was prepared by liquid

partitioning into water/methanol (polar fraction) and chlo-roform (non-polar fraction) as described earlier (Roessner

et al., 2000; Wagner et al., 2003). Metabolite samples

were derivatized by methoxyamination, using a 20 mg ml�1

solution of methoxyamine hydrochloride in pyridine,

and subsequent trimethylsilylation, with N-methyl-N-

1666 | Fuhrs et al. by guest on Septem

ber 29, 2015http://jxb.oxfordjournals.org/

Dow

nloaded from

(trimethylsilyl)-trifluoroacetamide (Fiehn et al., 2000;

Roessner et al., 2000). A C12, C15, C19, C22, C28, C32, and

C36 n-alkane mixture was used for the determination of

retention time indices (Wagner et al., 2003). Ribitol and

deuterated alanine were added for internal standardization.

Samples were analysed using GC-TOF-MS (ChromaTOF

software, Pegasus driver 1.61; LECO, http://www.leco.com).

Four sample types (6Mn and 6Si), each with fivereplicates, comprised an experimental data set of 20

chromatograms. The chromatograms and mass spectra were

evaluated using the TagFinder software (Luedemann et al.,

2008).

Sample preparation for the metabolite profiling of the

AWF was adapted to the respective volumes and metabolite

concentrations. In this case 200 ll of AWFH2O and

AWFNaCl were extracted to obtain a polar metabolitefraction, without further addition of water. The volume of

methanol/chloroform was reduced to 50% as were the

reagents for methoxyamination and silylation. Four sample

types (two Mn treatments, and two Si treatments), each

with four to five replications, in total 35 chromatograms,

were analysed as described above.

In parallel free phenols (in the following termed non-

polar apoplastic fraction) were extracted from AWFH2O andAWFNaCl. First AWF was alkalized with 0.5 N NaOH

(ratio 1:1) overnight. Afterwards samples were acidified by

adding 5 N HCl (ratio 0.1125:1). Phenols were then

extracted by shaking with diethylether (ratio 1:1). Samples

were then dried under nitrogen atmosphere and prepared

for GC-MS analysis as described for AWF. Four sample

types (two Mn treatments and two Si treatments), each with

five to six replications, resulted in 48 chromatograms, whichwere processed as described.

GC-MS metabolite profiles were processed after conversion

into NetCdf file format using the TagFinder (Luedemann

et al., 2008) and NIST05 software (http://www.nist.gov/srd/

mslist.htm). The mass spectral and retention index (RI)

collection of the Golm metabolome database (Kopka et al.,

2005; Schauer et al., 2005) was used for manually super-

vised metabolite identifcation. Yet non-identified metaboliccomponents were disregarded for the present study. Peak

height representing a mass specific arbitrary detector re-

sponse was used for screening the relative changes of

metabolite pools. The initial mass specific responses were

normalized by leaf fresh weight and ribitol recovery. AWF

metabolite profiles were normalized to ribitol recovery and

AWF total volume of partitioned polar (water/methanol)

and non-polar (chloroform) AWF fractions.

Statistical analysis of GC-MS profiles

Prior to statistical data assessment, response ratios were

calculated based on the mean response of each metabolicfeature from all samples of an experimental data set.

Response ratios were subsequently log10-transformed. In-

dependent component analysis (ICA) and missing value

substitution was as described earlier (Scholz et al., 2005).

ICA was carried out using the first five principal compo-

nents obtained from a set of manually identified metabolites

represented by at least three specific mass fragments each.

Basic calculations of relative changes in abundance of

specific metabolites due to Mn and Si treatment were made

with the Microsoft Excel 2000 software program and

respective embedded algorithms. For pairwise comparisons

thresholds of 2-fold change in pool size and P < 0.05

(t test,) were applied or levels of significance indicated,namely ***, **, and * representing P < 0.001, 0.01, and

0.05, respectively. Logarithmic transformation of response

ratios approximated the required Gaussian normal distribu-

tion of metabolite profiling data (Schaarschmidt et al.,

2007).

Statistical analysis of Mn and Si concentrations andapoplastic enzyme activities

Statistical analysis, if not mentioned otherwise, was carried

out using SAS Release v8.0 (SAS Institute, Cary, NC).

Results from analysis of variance are given according to

their level of significance as ***, **, and * for P < 0.001,

0.01, and 0.05, respectively. Pairwise comparisons were by

using Student’s t test.

Results

Exposing the plants to 50 lM Mn supply rapidly increasedthe Mn tissue concentration in the second oldest trifoliate

leaf over the 4 d treatment period (Fig. 1A). This led to

typical Mn toxicity symptoms (brown spots) after 2 d in-

creasing up to 70 spots cm�2 after 4 d of Mn treatment

(Fig. 1B). Silicon supply did not affect leaf Mn accumula-

tion (Fig. 1A). However, in contrast to plants cultivated

without Si, Si-treated plants developed only slight Mn

toxicity symptoms (2–5 spots cm�2) after 4 d of Mntreatment (Fig. 1B).

Since our previous work indicated a particular role of the

apoplast in the expression of Mn toxicity and Mn tolerance

in cowpea, our studies were focused on the AWF in

particular. In this study, the leaves were submitted to

a fractionated AWF extraction procedure yielding a free

water-soluble fraction (AWFH2O) and an ionically bound

NaCl-extractable (AWFNaCl) fraction. The Mn concentra-tion in the AWFH2O increased rapidly after 1 d of toxic Mn

supply and then it tended to decrease again (Fig. 2A).

Silicon application consistently enhanced the monomeric Si

concentration in the AWFH2O (Fig. 2B) compared with non

Si-treated plants, without consistently affecting the apo-

plastic Mn concentration (Fig. 2A). In the AWFNaCl, the

Mn concentration of the second trifoliate leaf steeply

increased after 1 d Mn treatment and remained stable ata higher level than in the AWFH2O (Fig. 2C). In Si-treated

plants, the Mn concentrations were slightly higher. Silicon

treatment enhanced the monomeric Si concentration (Fig.

2D), but with Mn treatment duration this difference

disappeared.

Apoplastic peroxidases and metabolites in manganese toxicity | 1667 by guest on Septem

ber 29, 2015http://jxb.oxfordjournals.org/

Dow

nloaded from

In order to demonstrate the capability of the PODisoenzymes to catalyse both H2O2-producing and -consum-

ing POD activities, AWFNaCl was separated by BN-PAGE

and PODs in-gel stained first for NADH-peroxidase fol-

lowed by staining for guaiacol-peroxidase activity (Fig. 3A,

B). Despite the quite low NADH-peroxidase activity stain-

ing intensity the gels revealed that each isoenzyme showedboth activities. Staining with guaiacol visualized major

isoenzymes more clearly: one isoenzyme smaller than P1

and four isoenzymes greater than P1, all with low activity

levels (Fig. 3B). After 6 d of Mn treatment, three additional

guaiacol-peroxidase bands appeared greater than the P6

Fig. 2. Effect of Mn treatment duration and Si supply on the Mn concentration (A, C) and the monomeric Si concentration (B, D) in the

water-soluble apoplastic fraction (A, B), and in the ionically bound apoplastic fraction (C, D) of the second oldest trifoliate leaves of the

Mn-sensitive cowpea cultivar TVu 91. After 2 weeks of preculture at 0.2 lM Mn, the Mn supply was increased to 50 lM for 4 d. Silicon

was supplied throughout plant culture. Results of the analysis of variance are given according to their level of significance as ***, **or * for

P <0.001, 0.01, or 0.05, respectively. Upper case and lower case letters indicate significant differences between Mn treatment duration

of –Si and +Si-treated plants, respectively, at P <0.05. An asterisk on top of the columns indicates siginificant differences between the Si

treatments for at least P <0.05 according to Tukey. Values are means 6SD with n¼16.

Fig. 1. Effect of Mn treatment duration and Si supply on (A) the Mn tissue concentration and (B) the density of brown spots of the

second oldest trifoliate leaves of the Mn-sensitive cowpea cultivar TVu 91. After 2 weeks of preculture at 0.2 lM Mn the Mn supply was

increased to 50 lM for 4 d. Silicon was supplied throughout plant culture. Results of the analysis of variance are given according to their

level of significance as ***, **or * for P < 0.001, 0.01, 0.05, respectively. Values are means 6SD with n¼16.

1668 | Fuhrs et al. by guest on Septem

ber 29, 2015http://jxb.oxfordjournals.org/

Dow

nloaded from

isoenzyme and one with a MW smaller than P1. One

isoenzyme with a MW greater than P1 disappeared owing

to elevated Mn supply. An extensive study of in-gel activity-stained BN gels loaded rigourously with the same protein

quantities comparing Mn treatments with and without Si

supply and differentiating between AWFH2O and AWFNaCl

proteins revealed that all isoenzymes were qualitatively

present in both Mn treatments, but elevated Mn supply led

to an increased abundance of isoenzymes P3 and P5,

especially in the water-soluble fractions (see Supplementary

Figs S1 and S2 at JXB online). In Mn-control plants Si-treatment did not affect the POD isoenzyme pattern. Silicon

delayed but not suppressed the Mn-mediated increase in the

number of POD isoenzymes in the AWFH2O (see Supple-

mentary Fig. S1 at JXB online).

Figure 3C shows a close-up of those POD isoenzymes

(clearly appearing after 4 d of Mn treatment), which were

chosen for further characterization after elution of the

proteins from the gels: P1, P3, P5, and P6, whereas P2 and

P4 were only sequenced. The eluted isoenzymes P3, P5, and P6,showed both NADH-peroxidase and guaiacol-peroxidase

activities (Fig. 4A, B). The specific activity was highest for

P6 followed by P5. The POD isoenzyme P1 had very little

guaiacol-peroxidase activity. The pH optimum for all

isoenzymes showing activity was consistently 6.5 for guaiacol-

peroxidase activity (Fig. 4A) and pH 5.5 for NADH-

peroxidase activity (Fig. 4B).

All marked POD activity-stained protein bands (Fig. 3)were cut; proteins were digested and analysed by liquid

chromatography-coupled mass spectrometry (LC-MS/MS).

MS/MS searches did not always lead to a positive identifi-

cation in cowpea (Vigna unguiculata) since its genome has

not yet been sequenced, but can lead to the identification of

peptides in related sequences of green plants (Viridiplantae)

downloaded from http://www.ncbi.nlm.nih.gov/sites/entrez.

Forty-four unique proteins were identified in the greenplants database. To estimate the false positive rate of

identification, a target-decoy database was performed (Elias

and Gygi, 2007), and no additional protein was identified in

reversed sequences, suggesting that our dataset contained

very few or no false-positive identifications. A list of all

resulting peptides, as well as their identities, is given as

supplementary data (see Supplementary Table S2 at JXB

online). Among these peptides, 11 peptides belonging toclass III peroxidases could be identified (Fig. 5). At least

three overlapping peptides provide evidence for at least

three distinct gene products. Three peptides with amino acid

substitutions were exclusively found in POD isoenzyme P1

when extracted with NaCl from Mn-treated plants (Figs 3,

5; see SupplementaryTable S2 at JXB online).

Since apoplastic NADH-peroxidase proved to react most

sensitively to toxic Mn supply and this enzyme has beenattributed a key role in the expression of Mn toxicity (Fecht-

Christoffers et al., 2006, 2007), the NADH-peroxidase

activity of the isoenzymes was further characterized for

interaction with different commercially available phenols

(Fig. 6) at the optimum pH identified above with p-coumaric

acid and Mn as a cofactors. Among the 10 phenols tested, p-

coumaric acid and vanillic acid proved to be the most

effective cofactors for all isoenzymes particularly at thehighest concentration level. Benzoic acid showed only little

activity at the higher concentrations even though the re-

sponse pattern was similar, whereas ferulic acid activated

NADH-peroxidase activity only at a lower concentration. All

other phenols did not induce NADH-peroxidase activity. As

shown above (Fig. 4A, B) the isoenzyme P6 showed by far

the highest activity.

The potential inhibitory effect of phenols on NADH-peroxidase activity was studied by adding eight phenols

to the reaction mixture and monitoring their effect on

Fig. 3. AWFNaCl-proteins of the second oldest trifoliate leaf of the

Mn-sensitive cultivar TVu 91 stained for (A) NADH-peroxidase and

(B) guaiacol-peroxidase activity after separation by BN-PAGE.

After preculture with 0.2 lM Mn (–Mn) for 14 d, plants received 50

lM (+Mn) Mn for 6 d. Fifty ll of concentrated AWFNaCl containing

ionically bound proteins (–Mn 60 lg, +Mn 112 lg) were loaded

onto the gels. Proteins were NBT-stained for NADH-peroxidase (A)

at pH 5.0 with 16 mM MnCl2, 1.66 mM p-coumaric acid, 0.625

mg ml�1 NBT, and 0.22 mM NADH. For guaiacol-peroxidase,

proteins were stained (B) in 18 mM guaiacol (in 9 mM Na2HPO4)

and 0.03% H2O2 at pH 6.0. Close up (C) shows marked

isoenzymes (P1, P3, P5, P6) that were chosen for elution and

further characterzation of pH optima and substrate specificity.

Apoplastic peroxidases and metabolites in manganese toxicity | 1669 by guest on Septem

ber 29, 2015http://jxb.oxfordjournals.org/

Dow

nloaded from

p-coumaric acid-stimulated enzyme activity (Fig. 7). Ben-

zoic acid and vanillic acid did not reduce the p-coumaric

acid-stimulated NADH-peroxidase activity and even en-hanced it. All other phenols inhibited NADH-peroxidase

activity by about 50% (ferulic and syringic acid) and by

>90% for the other phenols. This was true for all

isoenzymes.

Since metabolites were shown to affect apoplastic PODs

strongly (see above and Fecht-Christoffers et al., 2006), the

bulk-leaf metabolome was studied in a broad range

approach using GC-MS and independent component anal-yses (ICA) (Scholz et al., 2004). Applying ICA, sample

clusters were investigated according to the major variances

due to the treatment-induced qualitative and quantitative

changes of metabolite pools. This variance criterion was

augmented by subsequent pairwise or multiple probability-

based statistical significance testing.

In our factorial experimental designs both Mn and Si

treatment proved to be among the most important indepen-dent components (Fig. 8A) of our data sets resulting from

the bulk-leaf tissue. The analysis revealed that Mn (IC01)

and Si (IC04) treatments induced significant changes in the

metabolome. Silicon treatment clearly induced significant

conditional differences among the Mn control treatment but

only slight differences in Mn-treated plants. The Mn effect

was mainly caused by changes in the concentrations of amino

acids (serine, threonine, asparagine, aspartic acid), phenyl-alcohols (coniferylalcohol), organic acids (gluconic acid), and

sugar alcohols (sorbitol) as revealed by ICA loadings. The Si

effect was mainly due to differences in sugars (galactose) and

organic acids (gluconic acid).

In view of the particular role of the activity of apoplastic

peroxidases in Mn toxicity additionally the AWFH2O and the

AWFNaCl were subjected to a metabolomic analysis. The

ICA showed clear differences between the AWF fractions(IC01, Fig. 8B). Also, manganese treatment induced sepa-

rate clustering in both AWF fractions (IC02). In this

approach Si did not affect the sample clustering according

to treatment-mediated metabolite differences. As revealed

by ICA loadings, metabolites mainly responsible for thedifferential clustering of AWFH2O and AWFNaCl were

GABA, organic acids (malic acid, ribonic acid, gluconic

acid), amino acids (threonine), and sugars (xylose, eryth-

rose, fucose) among many currently unidentified metabo-

lites. The clustering according to the Mn treatment was

mainly caused by organic acids (maleic acid, malic acid,

nicotinic acid, itaconic acid), amino acids (threonine,

alanine), sugars (xylose, fructose, tagatose), and phenols (3-hydroxybenzoic acid).

Further fractionation of the leaf apoplastic metabolome

by an extraction method specifically yielding non-polar

metabolites revealed a clustering of samples according to

the infiltration solution, confirming the strong experimental

impact of the AWF fraction on the result (Fig. 8C).

Loadings derived from ICA showed that among other

currently unknown metabolites, mainly organic acids(fumaric acid, malic acid, succinic acid, citric acid, 3-

oxoglutaric acid) and phenylpropanoids (cis- and trans-

cinnamic acid, p-hydroxybenzoic acid) were responsible for

this clustering.

Quantification of relative changes between treatments

yielded five different phenols in this non-polar extract

(Table 1) among them ferulic acid, p-hydroxybenzoic acid,

and p-coumaric acid which had shown considerable inhibit-ing or enhancing effects, respectively, on in vitro NADH-

peroxidase activity. Ferulic acid and p-coumaric acid were

analytically separated into respective cis- and trans-isomers,

whereas in the in vitro NADH-peroxidase activity-enhancing/

inhibiting tests (Figs 6, 7) commercially available isomer

mixtures were used. Both ferulic acid isomers showed

a significant 2–4-fold reduction in abundance in Mn-treated

plants compared with control plants in the AWFH2O

fraction. A comparison of 6Si treatments revealed a signif-

icantly increased abundance of benzoic acid and of ferulic

Fig. 4. Determination of the pH optimum of (A) the guaiacol-peroxidase activity and of (B) the NADH-peroxidase activity of four POD

isoenzymes of the Mn-sensitive cowpea cultivar TVu 91. POD isoenzymes were eluted from BN gels that separated a mixture of AWFH2O

and AWFNaCl extracted from the second oldest trifoliate leaf of Mn-treated (4 d) and 6Si-treated (as described in the Materials and

methods) plants. Measurements were done in succinate buffer with pH values between 5.0 and 7.0 using 0.5 steps between the pH

values. Measuring solution (0.1 M succinate buffer) for the determination of NADH-peroxidase activity consisted of 16 mM MnCl2, 1.66

mM p-coumaric acid, and 0.22 mM NADH, measuring solution for guaiacol-peroxidase activity consisted of 18 mM guaiacol (in 90 mM

succinate buffer) and 0.03% H2O2.

1670 | Fuhrs et al. by guest on Septem

ber 29, 2015http://jxb.oxfordjournals.org/

Dow

nloaded from

Fig. 5. Alignment of determined and deduced amino acid (aa) sequences of peroxidases of various plant species and all 11 nano

LC-MS/MS-identified peroxidase peptide sequences from cowpea. Amino acid positions conserved in at least 50% of the sequences are

underlaid in grey. Asterisks (*) indicate the conserved distal haem-binding domain (I), the central conserved domain of unknown function

(II), and the proximal haem-binding domain. The eight cysteines (C1–C8) and the distal (Hd) and proximal (Hp) histidines are indicated,

too. Abbreviations: FBP1, French Bean Peroxidase 1 (acc no. AF149277); P49 (A.t.), POD isoenzyme 49 from Arabidopsis thaliana

(acc. no. O23237); PPOD from Populus ssp. (acc. no. AAX53172); VvPOD from Vitis vinfera (acc. no. CAO48839); VaPOD from Vigna

angularis (acc. no. BAA01950); P45 (A.t.) POD isoenzyme 45 from Arabidopsis thaliana (acc. no. Q96522); SoPOD from Spinacia

oleracea (acc. no. CAA71493); SiPOD from Sesamum indicum (acc. no. ABB89209), MsPOD from Medicago sativa (acc. no.

CAC38106); VuPOD, POD peptide sequences of Vigna unguiculata (this study).

Apoplastic peroxidases and metabolites in manganese toxicity | 1671 by guest on Septem

ber 29, 2015http://jxb.oxfordjournals.org/

Dow

nloaded from

acid isomers (more than 3-fold) in Si-treated plants only in

the AWFNaCl fraction. In Si-treated plants, high Mn supply

led to increased concentrations of benzoic acid in the

AWFH2O fraction and to decreased abundance of ferulic

acid compared with plants grown at low Mn supply. A com-parison of +Mn/+Si with +Mn/–Si (Mn toxicity-showing)

plants showed significantly decreased p-hydroxybenzoic

acid concentrations. A major, however not significant,

increase in abundance of cis-ferulic acid is indicated in the

+Mn/+Si plants not showing Mn toxicity symptoms.

NADH-peroxidase activity enhancing p-coumaric acid

showed no changes in abundance in each of the comparisons.

A three-factorial ANOVA showed benzoic acid, p-hydroxybenzoic acid, and ferulic acid to be significantly

affected by Mn (Table 2). Silicon treatment significantly

affected p-hydroxybenzoic acid and cis-ferulic acid. Highly

significant differences between the apoplastic fractions were

found for all identified phenylpropanoids except ferulic acid

and benzoic acid. Also, the infiltration solution had a clear

impact on p-hydroxybenzoic acid, p-coumaric acid, and

trans-sinapic acid. None of the two or three way inter-actions were significant (not presented).

Discussion

Effect of Mn and Si on apoplastic Mn fractions

Manganese is readily taken up by plants independent of the

Si supply, but the expression of toxicity symptoms was

suppressed by Si treatment (Fig. 1A, B) which is in line with

results previously published for cowpea (Horst et al., 1999;

Iwasaki et al., 2002a, b). This Si-enhanced Mn tolerance hasbeen explained entirely in cucumber (Rogalla and Romheld,

2002) or partly in cowpea (Iwasaki et al., 2002a, b) by

a reduction of the free Mn in the apoplast through enhanced

strong binding of Mn by the cell walls in Si-treated plants.

However, in the present study neither the AWFH2O (Fig. 2A)

nor the 5-fold higher AWFNaCl (Fig. 2C) Mn concentrations

differed clearly owing to Si treatment. This might be

explained by different growing conditions of the plants and

Mn extraction procedures. Nevertheless, this clearly shows

that, in cowpea, the expression of Mn toxicity cannot beexplained just on the basis of the free and exchangeable Mn

concentration in the leaf apoplast, in agreement with the

conclusion drawn by Iwasaki et al. (2002a, b). They

postulated a particular role of the monomeric Si in enhanc-

ing Mn tolerance. Indeed, also in our study the monomeric

Si concentration was consistently higher in Si-treated plants

in the AWFH2O (Fig. 2B) and initially also in the AWFNaCl

(Fig. 2D) fraction. The decreasing concentration of mono-meric Si with increasing Mn treatment duration in the latter

fraction possibly due to polymerization and/or strong

binding in the cell walls (incrustation) may explain why Si

treatment did not prevent but only delayed the formation of

brown spots (Fig. 1B) with extended Mn treatment duration.

Manganese and Si-induced changes ofperoxidase activities

All isoenzymes were shown to perform both reaction cycles

(Figs 3, 4). Mn treatment led to an increased abundanceof POD isoenzymes (Fig. 3; see Supplementary Fig. S1 at

JXB online; Fecht-Christoffers et al., 2003b) thus explaining

enhanced apoplastic POD activities (Fecht-Christoffers

et al., 2006). Silicon treatment only delayed but not

suppressed the Mn-mediated increased abundance of POD

isoenzymes (see Supplementary Fig. S1 at JXB online),

which is in line with the delayed but not prevented

development of Mn toxicity symptoms (Fig. 1B). Usinghigher protein loadings BN-PAGE separation of AWFH2O

and AWFNaCl protein did not reveal qualitative but only

quantitative differences in POD isoenzyme patterning

between the infiltration solutions, indicating that all

detected isoforms are principally water-soluble (see

Fig. 5. (continued)

1672 | Fuhrs et al. by guest on Septem

ber 29, 2015http://jxb.oxfordjournals.org/

Dow

nloaded from

Supplementary Fig. S2 at JXB online), even though a low

protein loading could lead to the opposite conclusion (Fig. 3;

see Supplementary Fig. S1 at JXB online). The results confirm

a particular role of PODs in the AWFH2O in the modulation of

Mn toxicity (Fecht-Christoffers et al., 2006, 2007).

Characterization of the identified peroxidases

The sequencing of the POD activity-showing 1D-BN pro-

tein bands P1 to P6 revealed that each band was composed

of more than one protein (see Supplementary Table S2 atJXB online) confirming BN/SDS-PAGE results previously

published by Fecht-Christoffers et al. (2003b). All bands led

to the identification of at least one peptide with high

sequence homology to peroxidases in the NCBI green plants

database. In total, 11 different peptides have been identified

belonging to the class III secretory peroxidase family

including sequences for the conserved so-called ‘domain II’

(Hiraga et al., 2001)/‘domain D’ (Delannoy et al., 2003)(Fig. 5, see Supplementary Table S2 at JXB online). Three

overlapping peptide sequences provide evidence for the

presence of at least three distinct genes encoding for class

III secretory peroxidases (Fig. 5). Three peptides with

amino acid substitutions (including the overlapping peptide

sequences: Fig. 5) were exclusively found in AWFNaCl-

extracted isoenzyme P1 from Mn-treated plants (Figs 3, 5;

see Supplementary Table S2 at JXB online) indicating

specific apoplastic binding properties.

As MS analyses did not result in complete PODsequences, one can only speculate about the total number

of distinct class III secretory peroxidases in Vigna unguicu-

lata. Based on in gel activity stainings, peroxidases of a wide

range of MW were detected (Fig. 3; see Supplementary Figs

S1 and S2 at JXB online). There are several possibilities

leading to such great differences in the MW of the

isoenzymes. (i) Class III peroxidases belong to a large

multigenic family even though they are distinct proteins(Passardi et al., 2004) with MWs ranging 28 kDa up to 60

kDa (Hiraga et al., 2001). (ii) A protein oligomer showing

peroxidase activity is conceivable, such as a peroxidase

dimer. (iii) Depending on the degree of N-glycolization, the

native MW may vary thus leading to changes in the MW in

the order of P3–P6 (Fig. 3). (iv) Other apoplastic proteins

than class III peroxidases might also have peroxidative

activity, i.e. oxidoreductase and/or auxin-binding (germin-like) proteins, even though the sequencing results did not

identify proteins that could perform a peroxidative reaction

(see Supplementary Table S2 at JXB online).

Fig. 6. Effect of different phenols on NADH-peroxidase activity of four POD isoenzymes. POD isoenzymes were eluted from BN-gels

(see Fig. 4 and Materials and methods). Measuring solution (0.1 M succinate buffer, pH 5.5) consisted of 16 mM MnCl2, 0.22 mM NADH,

and phenols (benzoic acid, p-coumaric acid, ferulic acid, caffeic acid, chlorogenic acid, gallic acid, protocatechuic acid, syringic acid,

and vanillic acid) in different concentrations (1.6 mM, 0.166 mM, 0.016 mM, and 0.0016 mM). Only the four displayed phenols induced

NADH-peroxidase activity. For the calculation of enzyme activities, extinction coefficients were adapted (see Supplementary Table S1 at

JXB online). Results are from two independent experiments including plant growth and protein separation.

Apoplastic peroxidases and metabolites in manganese toxicity | 1673 by guest on Septem

ber 29, 2015http://jxb.oxfordjournals.org/

Dow

nloaded from

The role of pH in controlling apoplastic PODisoenzyme activities

A pH optimum seems to be necessary for POD self-

protection (Olsen et al., 2003). In addition, the pH could be

an important regulatory factor for the relative performance

of either the peroxidative or the peroxidative–oxidative

reaction cycle of the enzyme. If an apoplastic pH of about5.0–6.0 as shown for Vicia faba (Muhling and Lauchli,

2000) is assumed, both POD cycles are expected to have

high activities within this range (Fig. 5), indicating that the

apoplastic pH is not decisive in regulating the relative

contribution of each reaction cycle in response to toxic Mn

supply. The determined pH optimum for both POD

activities is precisely in the range of the recommended pH

of the measuring solutions for POD activity determinationin vitro in studies investigating lignin formation (Karkonen

et al., 2002). However, in studies on the hypersensive stress

response to leaf pathogens, NADH-peroxidase-mediated

H2O2 production proved to be related to an alkalization of

the apoplast (Bolwell et al., 1995, 1998, 2001; Pignocchi and

Foyer, 2003) suggesting differences between biotic and

abiotic stress responses.

Fig. 7. Effect of combining different phenols with p-coumaric acid

as the control phenol on the induction capability for NADH-

peroxidase activity of four POD isoenzymes. POD isoenzymes

were eluted from BN gels (see Fig. 3 and Materials and methods).

Measuring solution (0.1 M succinate buffer, pH 5.5) consisted of

0.166 mM p-coumaric acid, 16 mM MnCl2, 0.22 mM NADH, and

0.0166 mM of one of the following phenols to examine interactions

between phenols: benzoic acid, ferulic acid, caffeic acid, chloro-

genic acid, gallic acid, protocatechuic acid, syringic acid, or vanillic

acid. Activities are expressed as relative values in relation to

activities when p-coumaric acid was applied alone (in the same

concentration). For the calculation of enzyme activities, extinction

coefficients were adapted (see Supplementary Table S1 at JXB

online). Results are from two independent experiments including

plant growth and protein separation.

Fig. 8. ICA plot of the GC-MS-accessible (A) bulk-leaf metab-

olome, (B) the polar AWF metabolites, and (C) the non-polar

metabolites extracted from the AWF. The second oldest trifoliate

leaf of the Mn-sensitive cultivar TVu 91 was tested for Mn and Si

effects. After 14 d of preculture with or without Si, plants received

50 lM Mn (+Mn) for 3 d or 0.2 lM Mn (–Mn) continuously. Bulk-

leaf, AWF- and non-polar apoplastic metabolites were extracted

(n¼5 and 6, respectively) as described in the Materials and

methods. ICA was conducted using MetaGeneAlyse at http://

metagenealyse.mpimp-golm.mpg.de.

1674 | Fuhrs et al. by guest on Septem

ber 29, 2015http://jxb.oxfordjournals.org/

Dow

nloaded from

The role of metabolites in controlling apoplastic PODisoenzyme activities: metabolite profiling

In a broad range metabolomic approach, it has been shown

that Mn toxicity induced changes in the bulk-leaf metab-

olome according to ICA (Fig. 8A; IC01) consistent with our

recent results showing that Mn toxicity also affects sym-

plastic reactions using a combined proteomic/transcriptomic

and physiological approach (Fuhrs et al., 2008). The

involvement of the symplast in Mn toxicity is in line withstudies using other plant species showing Mn toxicity-

induced reduced CO2 assimilation capacity (Gonzalez and

Lynch, 1997, 1999; Gonzalez et al., 1998, common bean;

Nable et al., 1988; Houtz et al., 1988, tobacco) accompanied

by reduced chlorophyll contents (Gonzalez and Lynch,

1999; Gonzalez et al., 1998, common bean; Moroni et al,

1991, wheat), and high Mn-accumulation rates in chloro-plasts (Lidon et al., 2004, rice). Our metabolomic approach

also showed that Si supply led to a particular clustering of

the total leaf metabolome as revealed by ICA (Fig. 8A;

IC04). This is in agreement with the work of Maksimovic

et al. (2007) on Si/Mn-toxicity interaction in cucumber who

concluded that Si supply modulates the phenol metabolism.

A closer investigation of the apoplastic metabolome using

AWFNaCl and AWFH2O revealed that the infiltration solu-tion (IC01; Fig. 8B) was the most important factor explain-

ing differences between the extracted metabolome fractions.

Manganese (IC02) but not Si treatment affected both AWF

metabolome fractions. The ICA loadings identified organic

acids, amino acids, and sugars to be responsible for Mn and

infiltration solution-related clusterings (Fig. 8B), whereas

phenolic compounds were unexpectedly low since Fecht-

Christoffers et al. (2006, 2007) reported a Mn-inducedchange in the apoplastic water-soluble phenol composition

(and at later toxicity stages even in phenol concentration)

using HPLC separation of leaf AWFH2O in cowpea (see

discussion below). However, GC-MS based metabolite pro-

filing typically covers mostly primary metabolites explaining

the relative low abundance of phenolic compounds.

To overcome this problem, an additional special AWF-

extraction procedure was applied yielding non-polar metab-olites. This resulted in clustering only according to the

infiltration solution (Fig. 8C; IC02, see discussion below).

ICA loadings revealed, in addition to organic acids, that

phenylpropanoids were mainly responsible for the cluster-

ing. Among other detected aromatic compounds, ferulic

acid was identified as a clearly Mn and Si-affected phenol

(Tables 1, 2; see discussion below).

Table 1. Identified phenols (GC-MS) in the non-polar fraction of the leaf AWF recovered after infiltration with H2O or NaCl

Displayed are the relative pool-size changes of each phenol calculated on the basis of response ratios. The effects of these phenols on theNADH-peroxidase activity (see Figs 6 and 7) of apoplastic peroxidase isoenzymes are also shown. After 14 d of preculture, 6Si-treated plantsof the Mn-sensitive cowpea cv. TVu91 received 50 lM Mn for 3 d or 0.2 lM Mn continuously. Statistical testing of changes in metaboliteabundance were calculated using log10-transformed response ratios. An asterisk denotes significant differences at least at P <0.05 (n¼6),respectively (t test).

Detected metabolitesNADH-

+Mn/–Mn +Si/–Si +Mn+Si/–Mn+Si +Mn+Si/+Mn–Si Effect of phenol onperoxidase activitya

AWFH2O AWFNaCl AWFH2O AWFNaCl AWF AWFNaCl AWFH2O AWFNaCl

Benzoic acid 1.41c 1.35 0.91 1.32* 1.49* 1.14 0.97 1.12 weak induction/no inhibition

p-Hydroxybenzoic acidb 1.47 1.61 0.87 0.65 1.03 1.23 0.61* 0.50* No induction/50% inhibition

cis-p-Coumaric acid 1.00 0.81 0.95 1.13 1.06 0.62 1.00 0.86 Strong induction

cis-Ferulic acid 0.24* 0.30* 1.03 3.77* 0.70 0.35* 2.96 4.37 Weak induction/50% inhibition

trans-p-Coumaric acid 0.88 0.82 0.84 0.95 0.97 0.59 0.92 0.68 Strong induction

trans-Ferulic acid 0.44* 2.31 1.34 3.61* 0.50 0.37* 1.52 0.57 Weak induction/50% inhibition

trans-Sinapic acid 2.39 0.81 n.d.+ 0.72 n.d.++ 0.90 n.d.++ 0.80 Not examined

a from Figs. 6 and 7.b After identification of p-hydroxybenzoic acid, this phenol was additionally tested with respect to NADH-peroxidase activity. In addition to the

50% inhibitory effect it showed no induction capability for NADH-peroxidase activity for each isoenzyme tested.c Numbers are calculated ratios of the response ratios (not log10-transformed) within the individual comparison. ANOVA did not reveal

a significant Mn3Si interaction.+,++ were not detected (n.d.) in +Si and +Mn+Si treatments, respectively.

Table 2. Identified phenols (GC-MS) in the non-polar fraction of

the leaf AWF recovered after infiltration with H2O or NaCl (Inf.).

Displayed are the p-values derived from analysis of variance basedon log10-transformed response ratios (n¼6). For the effects of thesephenols on the NADH-peroxidase activity of apoplastic peroxidaseisoenzymes see Figs 6 and 7 as well as Table 1. After 14 d ofpreculture, 6Si-treated plants of the Mn-sensitive cowpea cultivarTVu 91 received 50 lM Mn for 3 d or 0.2 lM Mn continuously.

Metabolite Mn Si Inf.

Benzoic acid 0.0032 0.2786 0.2615

4-Hydroxybenzoic acida 0.0093 <0.0001 <0.0001

cis-4-Hydroxycinnamic acid 0.4236 0.5636 <0.0001

cis-Ferulic acid <0.0001 0.0012 0.4433

trans-4-Hydroxycinnamic acid 0.1269 0.1057 <0.0001

trans-Ferulic acid 0.0129 0.2470 0.3870

trans-Sinapic acid 0.4671 0.1685 0.0039

a After identification of p-hydroxybenzoic acid, this phenol wasadditionally tested with respect to NADH-peroxidase activity. Inaddition to the 50% inhibitory effect, it showed no induction capabilityfor NADH-peroxidase activity for each isoenzyme tested.

Apoplastic peroxidases and metabolites in manganese toxicity | 1675 by guest on Septem

ber 29, 2015http://jxb.oxfordjournals.org/

Dow

nloaded from

Overall, the broad-range metabolite profiling in the bulk-

leaf extract (Fig. 8A, ICA01) and the AWF (Fig. 8B;

ICA02) revealed a clear difference related to the Mn

treatment. The Si effect was less clearly expressed. A

preliminary metabolite-specific evaluation of the metabo-

lites indicates alterations of metabolic pathways mainly

related to organic acids, amino acids, and sugars/sugar

alcohols. A detailed evaluation and discussion of thequalitative changes in polar apoplastic metabolites is

beyond the scope of this paper and will be subject of

a subsequent paper.

The role of phenols in controlling apoplastic NADH-peroxidase activity

Analysing the AWFH2O using HPLC, Fecht-Christoffers

et al. (2006) separated water-soluble phenols in the apo-

plast. A Mn treatment not only increased the peak size but

also led to at least two additional peaks, which supported

their conclusion that the presence of phenols in the apoplast

is decisive for the expression of Mn toxicity/Mn tolerance incowpea leaf tissue. However, they failed to identify the

phenols. Our gas chromatography–mass spectrometry ap-

proach allowed us to identify five phenols. However, the

method does not allow absolute concentrations to be

determined but only relative treatment-related concentra-

tion changes. Also, it was not possible to identify most

phenols directly in the AWF. Therefore, the aqueous AWF

was extracted with diethylether which led to a concentrationof the phenols but at the same time only yielded non-polar

metabolites. Thus, the applied technique did not allow us to

identify and quantify all the phenols present in the apoplast

which is a major focus of ongoing research. Nevertheless,

among the phenols identified (Tables 1, 2), four were found

which had been tested for their effect on NADH-peroxidase

activity in vitro. Only p-coumaric acid had a strong activity-

enhancing effect. Ferulic acid and p-hydroxybenzoic acidhad only a weak or lacking stimulating effect, but a strong

inhibiting effect when combined with p-coumaric acid.

Benzoic acid only weakly enhanced and did not inhibit

NADH-peroxidase activity (Figs 6, 7; Table 1).

The three-factorial analysis of variance of the treatment-

induced changes in the abundance of the phenols (Table 2)

revealed that Mn treatment significantly affected the concen-

trations of benzoic, p-hydroxybenzoic and, most clearly,ferulic acid, whereas Si treatment affected p-hydroxybenzoic

and again most clearly cis-ferulic acid. Looking at the

comparison of means of the treatment-specific relative pool-

size changes of the individual phenols (Table 1), it appears

that the change in the concentration in the apoplast of ferulic

acid particularly plays a key role in the expression of Mn

toxicity symptoms: a reduction of the concentration leading

to a reduced inhibition of NADH-peroxidase activity ischaracteristic for leaves showing Mn toxicity symptoms

(+Mn/–SI), while Mn-tolerant leaf tissue (–Mn/+Si; +Mn/

+Si) is characterized by an enhanced accumulation. The

constitutive effect of Si on an enhanced abundance of ferulic

acid seems to be stong enough to counteract the Mn-induced

reducing effect (compare +Mn +Si/–Mn +Si, Table 1). Also,

it appears that Si affects the phenol concentration more in

the AWFNaCl (as indicated by the high infiltration solution

effect on the phenols in Table 2) than in the AWFH2O

corroborating results demonstrating Si-mediated changes of

apoplastic Mn-binding properties (Iwasaki et al., 2002a;

Rogalla and Romheld, 2002). However, ferulic acid and

benzoic acid, in particular, were not affected by the in-filtration solution, indicating specific apoplastic binding

properties in the apoplast for each phenol regardless of Si

nutrition (Table 1). The Si-induced significantly higher

abundance of benzoic acid might be of minor importance,

given the rather weak NADH-peroxidase activity-enhancing

effect (Fig. 8). However, the lowered concentration of

NADH-peroxidase activity-inhibiting p-hydroxybenzoic acid

in the presence of Si at high Mn supply is not in line with theabove expressed line of thinking. Thus it appears a more

detailed and quantitative investigation of the phenols present

in the leaf apoplast is necessary to understand Mn toxicity

and Mn tolerance fully.

In conclusion, the results presented here confirm the

hypothesized role of apoplastic NADH-peroxidase and its

activity-modulating phenols in Mn toxicity and Si-enhanced

Mn tolerance. Isoenzyme BN gel-profiling of POD enzymesand their characterization after elution from the gels, and

metabolite profiling of the bulk-leaf and the AWF appear to

be powerful tools in enhancing the physiological and

molecular understanding of Mn toxicity and Mn tolerance.

Supplementary data

Supplementary data can be found at JXB online.

Supplementary Fig. S1. 1D BN-PAGE resolution of

AWFH2O and AWFNaCl proteins (16 lg) after 0 and 4 d of

Mn treatment of 6Si-treated plants of the Mn-sensitive

cowpea cultivar TVu 91.

Supplementary Fig. S2. 1D BN-PAGE resolution of

AWFH2O and AWFNaCl proteins (180 lg) after 0 d and 4 dof Mn treatment of the Mn-sensitive cowpea cultivar

TVu 91.

Supplementary Table S1. Extinction coefficients for the

calculation of NADH-peroxidase activities of different POD

isoenzymes supplied with different phenols in changing

concentrations as shown in Figs 4, 6, and 7.

Supplementary Table S2. Peptide sequences of apoplastic

leaf proteins sequenced with LC-MS/MS.

Acknowledgements

This work was supported by the Deutsche Forschungsge-

meinschaft (grants HO 931-17, HO 931-18/1).

References

Bolwell GP, Butt VS, Davies DR, Zimmerlin A. 1995. The origin of

the oxidative burst in plants. Free Radical Research 23, 517–532.

1676 | Fuhrs et al. by guest on Septem

ber 29, 2015http://jxb.oxfordjournals.org/

Dow

nloaded from

Bolwell GP, Davies DR, Gerrish C, Auh C-K, Murphy TM. 1998.

Comparative biochemistry of the Oxidative Burst prodiced by rose and

French Bean cells reveals two distinct mechanisms. Plant Physiology

116, 1379–1385.

Bolwell GP, Page A, Pislewska M, Wojtaszek P. 2001. Pathogenic

infection and the oxidative defences in plant apoplast. Protoplasma

217, 20–32.

Bradford MM. 1976. A rapid and sensitive method for the quantita-

tion of microgram quantities of protein utilizing the principle of protein–

dye binding. Analytical Biochemistry 72, 248–254.

Delannoy E, Jalloul A, Assigbetse K, Marmey P, Geiger JP,

Lherminier J, Daniel JF, Martinez C, Nicole M. 2003. Activity of

class III peroxidases in the defense of cotton to bacterial blight.

Molecular Plant–Microbe Interactions 16, 1030–1038.

Elias JE, Gygi SP. 2007. Target-decoy search strategy for increased

confidence in large-scale protein identifications by mass spectrometry.

Nature Methods 4, 207–214.

Epstein E. 1999. Silicon. Annual Review of Plant Physiolgy and Plant

Molecular Biology 50, 641–664.

Fecht-Christoffers MM, Braun H-P, Lemaitre-Guillier C,

VanDorsselear A, Horst WJ. 2003b. Effect of Mn toxicity on the

proteome of the leaf apoplast in cowpea. Plant Physiology

133, 1935–1946.

Fecht-Christoffers MM, Fuhrs H, Braun H-P, Horst WJ. 2006.

The role of hydrogen peroxide-producing and hydrogen peroxide-

consuming peroxidases in the leaf apoplast of cowpea in manganese

tolerance. Plant Physiology 140, 1451–1463.

Fecht-Christoffers MM, Maier P, Horst WJ. 2003a. Apoplastic

peroxidase and ascorbate are involved in manganese toxicity

and tolerance of Vigna unguiculata. Physiologia Plantarum 117,

237–244.

Fecht-Christoffers MM, Maier P, Iwasaki K, Braun H-P,

Horst WJ. 2007. The role of the leaf apoplast in manganese toxicity

and tolerance in cowpea (Vigna unguiculata L. Walp). In: Sattelmacher

B, Horst WJ, eds. The apoplast of higher plants: compartment of

storage, transport, and reactions. Dordrecht, The Netherlands:

Springer, 307–322.

Fiehn O, Kopka J, Trethewey RN, Willmitzer L. 2000. Identification

of uncommon plant metabolites based on calculation of elemental

compositions using gas chromatography and quadrupole mass

spectrometry. Analytical Chemistry 72, 3573–3580.

Fuhrs H, Hartwig M, Molina LEB, Heintz D, Van Dorsselaer A,

Braun H-P, Horst WJ. 2008. Early manganese-toxicity response in

Vigna unguiculata L.: a proteomic and transcriptomic study. Proteo-

mics 8, 149–159.

Gonzalez A, Lynch JP. 1997. Effects of manganese toxicity on leaf

CO2 assimilation of contrating common bean genotypes. Physiologia

Plantarum 101, 872–880.

Gonzalez A, Lynch JP. 1999. Subcellular and tissue Mn compart-

mentation in bean leaves under Mn toxicity stress. Australian Journal

of Plant Physiology 26, 811–822.

Gonzalez A, Steffen KL, Lynch JP. 1998. Light and excess

manganese: Implications for oxidative stress in common bean. Plant

Physiology 118, 493–504.

Halliwell B. 1978. Lignin synthesis: the generation of hydrogen

peroxide and superoxide by horseradish peroxidase and its stimulation

by manganese (II) and phenols. Planta 140, 81–88.

Hiraga S, Sasaki K, Ito H, Ohashi Y, Matsui H. 2001. A large family

of class III plant peroxidases. Plant and Cell Physiology 42, 462–468.

Horst WJ. 1980. Genotypische Unterschiede in der Mangan-Toleranz

von Cowpea (Vigna unguiculata). Angewandte Botanik 54, 377–392.

Horst WJ. 1982. Quick screening of cowpea genotypes for manga-

nese tolerance during vegetative and reproductive growth. Zeitschrift

fur Pflanzenernahrung und Bodenkunde 145, 423–425.

Horst WJ. 1983. Factors responsible for genotypic manganese

tolerance in cowpea (Vigna unguiculata). Plant and Soil 72,

213–218.

Horst WJ. 1988. The physiology of Mn toxicity. In: Webb MJ, Nable

RO, Graham RD, Hannam RJ, eds. Manganese in soil and plants.

Dodrecht/Boston/London: Kluwer Academic Publishers, 175–188.

Horst WJ, Fecht M, Naumann A, Wissemeier AH, Maier P. 1999.

Physiology of manganese toxicity and tolerance in Vigna unguiculata

(L.) Walp. Journal of Plant Nutrition and Soil Science 162, 263–274.

Horst WJ, Marschner H. 1978a. Effect of silicon on manganese

tolerance of bean plants (Phaseolus vulgaris L.). Plant and Soil

50, 287–303.

Horst WJ, Marschner H. 1978b. Symptome von Manganuberschuss

bei Bohnen (Phaseolus vulgaris). Zeitschrift fur Pflanzenernahrung und

Bodenkunde 141, 129–142.

Houtz RL, Nable RO, Cheniae GM. 1988. Evidence for effects on

the in vivo activity of ribulose-bisphosphate carboxylase/oxygenase

during development of Mn toxicity in tobacco. Plant Physiology

86, 1143–1149.

Iwasaki K, Maier P, Fecht M, Horst WJ. 2002a. Effects of silicon

supply on apoplastic manganese concentrations in leaves and their

relation to manganese tolerance in cowpea (Vigna unguiculata (L.)

Walp.). Plant and Soil 238, 281–288.

Iwasaki K, Maier P, Fecht M, Horst WJ. 2002b. Leaf apoplastic

silicon enhances manganese tolerance of cowpea (Vigna unguiculata).

Journal of Plant Physiology 159, 167–173.

Iwasaki K, Matsumura A. 1999. Effect of silicon on alleviation of

manganese toxicity in pumpkin (Cucurbita moschata Duch cv.

Shintosa). Soil Science and Plant Nutrition 45, 909–920.

Jansch L, Kruft V, Schmitz UK, Braun HP. 1996. New insights into

the composition, molecular mass and stoichiometry of the protein

complexes of plant mitochondria. The Plant Journal 9, 357–368.

Karkonen A, Koutaniemi S, Mustonen M, Syrjanen K, Brunow G,

Kilpelainen I, Teeri TH, Simola LK. 2002. Lignification related

enzymes in Picea abies suspension cultures. Physiologia Plantarum

114, 343–353.

Kopka J, Schauer N, Krueger S, et al. 2005. GMD@CSB.DB: the

Golm metabolome database. Bioinformatics 21, 1635–1638.

Lidon FC, Barreiro MG, Ramalho JC. 2004. Manganese accumu-

lation in rice: implications for photosynthetic functioning. Journal of

Plant Physiology 161, 1235–1244.

Luedemann A, Strassburg K, Erban A, Kopka J. 2008. TagFinder

for the quantitative analysis of gas chromatography-mass spectrometry

Apoplastic peroxidases and metabolites in manganese toxicity | 1677 by guest on Septem

ber 29, 2015http://jxb.oxfordjournals.org/

Dow

nloaded from

(GC-MS) based metabolite profiling experiments. Bioinformatics

24, 732–737.

Maksimovic JD, Bogdanovic J, Maksimovic V, Nikolic M. 2007.

Silicon modulates the metabolism and utilization of phenolic com-

pounds in cucumber (Cucumis sativus L.) grown at excess manga-

nese. Journal of Plant Nutrition and Soil Science 170, 739–744.

Marschner H. 1995. Mineral nutrition in higher plants, 2nd edn.

London, UK: Academic Press.

Moroni JS, Briggs KG, Taylor GJ. 1991. Chlorophyll content and

leaf elongation rate in wheat seedlings as a measure of manganese

tolerance. Plant and Soil 136, 1–9.

Muhling KH, Lauchli A. 2000. Light-induced pH and K+ changes in

the apoplast of intact leaves. Planta 212, 9–15.

Nable RO, Houtz RL, Cheniae GM. 1988. Early inhibition of

photosynthesis during development of Mn toxicity in tobacco. Plant

Physiology 86, 1136–1142.

Olsen LF, Hauser MJB, Kummer U. 2003. Mechanism of protection

of peroxidase activity by oscillatory dynamics. European Journal of

Biochemistry 270, 2796–2804.

Passardi F, Cosio C, Penel C, Dunand C. 2005. Peroxidases have

more functions than a Swiss army knife. Plant Cell Reports 24,

255–265.

Passardi F, Longet D, Penel C, Dunand C. 2004. The class III

peroxidase multigenic family in rice and its evolution in land plants.

Phytochemistry 65, 1879–1893.

Pignocchi C, Foyer CH. 2003. Apoplastic ascorbate metabolism and

its role in the regulation of cell signalling. Current Opinion in Plant

Biology 133, 443–447.

Roessner U, Wagner C, Kopka J, Trethewey RN, Willmitzer L.

2000. Simultaneous analysis of metabolites in potato tuber by

gas chromatography–mass spectrometry. The Plant Journal 23,

131–142.

Rogalla H, Romheld V. 2002. Role of leaf apoplast in silicon-

mediated manganese tolerance of Cucumis sativus L. Plant, Cell and

Environment 25, 549–555.

Schaarschmidt S, Kopka J, Ludwig-Muller J, Hause B. 2007.

Regulation of arbuscular mycorrhization by apoplastic invertases:

enhanced invertase activity in the leaf apoplast affects the symbiontic

interaction. The Plant Journal 51, 390–405.

Schauer N, Steinhauser D, Strelkov S, et al. 2005. GC-MS libraries

for the rapid identification of metabolites in complex biological

samples. FEBS Letters 579, 1332–1337.

Scholz M, Gatzek S, Sterling A, Fiehn O, Selbig J. 2004.

Metabolite fingerprinting: detecting biological features by

independent component analysis. Bioinformatics 20,

2447–2454.

Scholz M, Kaplan F, Guy CL, Kopka J, Selbig J. 2005. Non-linear

PCA: a missing data approach. Bioinformatics 21, 3887–3895.

Shi Q, Bao Z, Zhu Z, He Y, Qian Q, Yu J. 2005. Silicon-mediated

alleviation of Mn toxicity in Cucumis sativus in relation to activities of

superoxide dismutase and ascorbate peroxidase. Phytochemistry

66, 1551–1559.

Wagner C, Sefkow M, Kopka J. 2003. Construction and application

of a mass spectral and retention time index database generated

from plant GC/EI-TOF-MS metabolite profiles. Phytochemistry

62, 887–900.

Wehrhahn W, Braun HP. 2002. Biochemical dissection of the

mitochondrial proteome from Arabidopsis thaliana by three-

dimensional gel electrophoresis. Electrophoresis 23, 640–646.

Wissemeier AH, Horst WJ. 1992. Effect of light intensity on

manganese toxicity symptoms and callose formation in cowpea (Vigna

unguiculata (L.) Walp.). Plant and Soil 143, 299–309.

Yamazaki I, Piette LH. 1963. The mechanism of aerobic oxidase

reaction catalysed by peroxidase. Biochimica et Biophysica Acta

77, 47–64.

1678 | Fuhrs et al. by guest on Septem

ber 29, 2015http://jxb.oxfordjournals.org/

Dow

nloaded from